Introduction by W.W. Skeat, continued

{xxviii}
§ 14. The Metre. In an article on ‘The Scansion of English Poetry,’ printed in the Phil. Soc. Transactions for 1898, I proposed a natural method of scansion which is much more suitable (in my opinion) for the scansion of our native poetry than the usual one which applies the inappropriate Greek terms ‘iamb’ and ‘trochee’ to a Teutonic language. A description of this method is also given in my edition of Chaucer, vol. vi. p. lxxxiv; and I beg leave to repeat it here, as it enables us to see that the lines in Havelok are of sixteen distinct types, and saves a great deal of discussion. I have only to add that the system here given is by no means original, but was proposed and very fully illustrated in an anonymous tractate entitled ‘Accent and Rhythm, explained by the law of Monopressures. Part I. Edinburgh, 1888.’

§ 15. Speech-waves. In English, accent or stress plays a very important part; and for this reason we may consider English speech as made up of a succession of utterances which form, as it were, speech-waves, in which each wave, due to a jet of breath, contains a strong, i.e. a stressed syllable; and this strong syllable may either stand alone (as in the word tóne); or may be preceded or followed by a weak (or unstressed) syllable (as in the words ascént, cádence); or may even be both preceded and followed by a weak syllable during the emission of the same jet of breath (as in the word exténsion).

This amounts to saying that the words light, alight, lighted, and alighted can all be produced in a single speech-wave. But if a word has two stresses, it requires two impulses to utter it, and really contains two speech-waves. Such words are very common; as cónque-ròr, amál-gamàte, &c., in which {xxix} one stress is stronger than the other; and many English words require three speech-waves, as insòl-ubíli-tỳ; or even four, as ìn-combùsti-bíli-tỳ. It often happens that the secondary stresses are very slight; but the ear should be trained to detect their existence. In order to denote the exact effect produced by the pronunciation of the words light, alight, lighted, and alighted, we may use the terms ‘tone,’ ‘ascent,’ ‘cadence,’ and ‘extension’ already mentioned; which may further, for brevity, be denoted by the initial letters t, a, c, e; or (in roman type) t, a, c, e.

§ 16. The sixteen types of Verse. As it is often desirable to employ other symbols which appeal to the eye more directly, I shall also use the small letter ‘b’ to denote an unstressed syllable, and the capital letter ‘A’ to denote a stressed one. Then, as each line contains four stressed syllables, the general scheme for its scansion may be roughly denoted by ‘bAbAbAbA.’ But it frequently happens that the initial weak syllable is lacking, giving a line which may be denoted by ‘AbAbAbA.’ Moreover, the final rhyme is often a double or feminine one, in which case we have lines that may be represented by ‘bAbAbAbAb’; or by ‘AbAbAbAb.’

It is usual to analyse the scansion no further, and to be satisfied with chopping up the lines into ‘ feet,’ as in Latin prosody, which really deals with ‘short’ and ‘long’ syllables. But the method above indicated enables us to go much more closely into the structure of the verse, and to distinguish no less than eight varieties of each of the above types; or, if we neglect the subordinate and accidental change from a masculine rhyme to a feminine one, we can still detect eight varieties of each of the principal types above named, viz. bAbAbAbA and AbAbAbA, according to the various ways in which the syllables can be arranged in groups or {xxx} speech-waves. Hence the sixteen primary types can be expressed as follows:—

These eight types may be still more briefly denoted, as explained above, by the following formulae:—

By the simple expedient of dropping the initial weak syllable, we obtain eight new types, as follows:—

The shorter formulae are:—

These schemes presuppose the lines to have single or ‘masculine’ rhymes, as in ll. 17, 18, which end with do and to. But, as a matter of fact, a large proportion of the lines are furnished with double or ‘feminine’ rhymes, as in ll. 3 and 4, which end with tel-lë, duel-lë; for the final e is. invariably pronounced, at the end of a line, as a distinct syllable. But this does not really alter the type of the verse. Line 3 is to be read thus: Óf.a tál’.ich yóu.wil télle. This belongs to type 9, with the substitution of an ‘extension’ for the final ‘ascent.’ We may denote it by 9b; where the b reminds us that the rhyme is double.

§ 17. Elision and Contraction. It must be remembered that, in Middle English verse, elision of a final vowel before a word beginning with a vowel is extremely common. {xxxi} Further, in the copy of Havelok here printed, it is tolerably certain that the scribe has frequently written final -en in the place of final -e, and this -en is therefore to be elided in the same way. Thus, in l. 12, the word mowen stands for mowe, which becomes mow’ before the following y-. Or, if we wish to preserve the n, we may suppress the e, and call it mow’n. But it is dissyllabic in l. 11, where followed by a consonant.

A few other rules are observed, as in other M.E. writings; amongst which the following may be noted.

1. A final -e (or often -en) is elided not only before a vowel, but before any one of the words beginning with h in the following list, viz, he, his, him, hire (her), here (their), hem, hath, hadde, haue, how, heer. Thus in l. 18, comen (for come) is merely com’, before the word him.

2. The endings -es, -ed, and -en or -e (before a consonant) usually count as a distinct syllable. Thus we have burw-es, 55, Eng-e-lond-es, 63; but wreieres and robberes in l. 39 should be wreiers and robbers, because the accent does not fall upon the syllable that precedes -es; cf. feter’s, 82. In l. 5 we find the full form i-mak-ed; but in ll. 23, 58, maked is probably miswritten for mād, as in l. 1953; cf. mad’ (he made) in ll. 38, 39, 41.

3. Many words that have a medial e are familiarly contracted; thus eueri is ev’ri, 8; euere is ev’re, 17, 88, and is often written eure on this account, as in 424; cf. neure, i.e. nevre, in 80, 108. Hauelok is often Hav’lok, as in 5, 7; oueral is ov’ral, 38, 54; hauede is hav’de (A.S. hæfde), 90, 98; pouere is pov’re, 58, written poure, 101; louede is lov’de, 30, and even lov’d’, 71. Such instances are numerous.

4. The final -e, properly a distinct syllable, has, as usual, a true grammatical or etymological significance. Thus mal-e, 48, represents O.F. mal-e, O.H.G. mal-ha; and gom-e, trom-e, 7, 8, represent A.S. gum-a, trum-a. God-e, 1, 28, {xxxii} is the pl. of the adjective good; cf. riht-wis-e, 37. In 87, þe best-e represents the definite form of the adjective. The adverbial suffix -e appears in hey-e, 43, þann-e, 51; bald-elik-e, 53, son-e, 81. The infinitive of the verb sometimes ends in -en, as in drink-en, 15; but usually in -e, as in fell-e, 3, duel-le 4, fall-e, 39, bynd-e, 41. Even when written with -en, it is better to substitute -e, as in biginne, i.e. biginn’, 21; heng’, 43; lurk’, crep’, 68. The pt. t. sing. of weak verbs ends in -e; as þurt-e, 10; miht-e, 42; &c. In l. 22, yeu-e (yev-e) is the subjunctive form, used as an imperative; cf. bor-e, pt. t. s. subj., 45. In these and other similar cases, the rules are much the same as those which explain the scansion of Chaucer, and need not be enlarged upon. In a few cases, the final -e seems to have been suppressed, as in yed’ (before a consonant), 6; where ‘ẹ’ signifies a mute ‘e.’

§ 18. The Caesural Pause. Most of the lines readily admit of a caesural pause, which naturally occurs at the end of the second speech-wave. This often admits of the easy introduction of an extra weak syllable at this point. Thus, in l. 6, we find was lit-el, an ‘extension’ (e), where the metre would otherwise admit only of an ‘ascent’ (a). And, in l. 2, we find (after the pause) and ál-lè (e), where the metre would otherwise admit only of a ‘cadence’ (c). This is a familiar phenomenon in English poetry, and such lines may be denoted by affixing e (extended group) to the number of the type. Thus l. 2 is really of the type 16e, i.e. (ccct); but with (e) for (c) in the third sound-group. Similarly, l. 6 is really of the type 1be, i.e. (aaaa); but with a feminine rhyme (§ 16), and with (e) for (a) in the second sound-group. It thus becomes perfectly easy to denote the precise type to which any given line belongs.

§ 19. Examples of Scansion. A few examples will make this clearer.

{xxxiii}
1 (aaaa). Ex. Ne-fúnd’ he-nón pat-déd’ hem-shám; 56. To the same type we may reduce ll. 68 (with two elisions of final -en, written for -e); 216; 288; 479; &c.

1b (aaae); with feminine rhyme. Ex. Fil-mé¹ a-cúpp’ of-fúl god-ále; 14. So also ll. 49; 60 (with Michel, really a ‘cadence,’ in the first sound-group, by the same licence as in 14); 64; 91 (read he ne as hé n’); 92; 112; 119; 130; 157; 468; &c.

¹ Or rather, Fíl-me (c); by a common licence.

Line 6 belongs to the type 1be (§ 18).

2 (aaet). Ex. That-hé ne-wér’n to-sórwe bróuht; 57. So also ll. 74; 108; 115; 149 (taking nouht-búten together); 192; 289.

2b (aaec). Ex. He-wás ful-gód in-év’ri tróme; 8. So also 4 (see note); 27; 28; 65; 86; 99 (taking non-só-god together); 133; 188.

2e (aeet). Ex. He-déd’ him-sóne ǁ to-háuen ríht; 78². So also 144.

² Or: hem sóne t’háuen riht; type 4.

2be (aeec). Ex. Ne-hé ne-móuhte ǁ no-lýþe géte; 147.

3 (aeta). Ex. The-rým is-máked óf Hav’lók; 23³. So also 44; 61; 107; 145 (read hav’ as a monosyllable).

³ Or: is-máked ǁ of-Háve.lok; type 2e (aeet).

3b (aete). Ex. The-tál’ of-Háv’lok ís y-máked; 5. So also 9; 15; 16 (read all’); 25; 39; 41; 87; 94 (And-óþ’r he-réft’-him), where the final -er in óþer is practically elided (or much slurred over) before the following he; 138 (to which a similar remark applies). This is a favourite and flowing type.

4 (aect). Ex. The-kíng was-hóten Áthel.wóld; 106; 137 (with Áfter in the first group).

{xxxiv}
4b (aecc). Ex. And-ál for-hís-e gód-e wórkes; 34. Cf. 81; 90 (ne-háv’d’-he név’re); 100; 128; 132 (Ne-wóld’).

5 (etaa). Ex. He-lóv’de Gód with-ál his-míht; 35. Cf. 36; 47; 89; 109; 124.

5b (etae). Ex. And-háted hém so-mán doth-gálle; 40. Cf. 46; 221; 230; 277.

6 (etet). Ex. And-héye heng’ on-gálwe tré; 43. Cf. 117.

6b (etec). Ex. He-déde mak’ and-fúl-wel hólden; 29. Cf. 30; 37; 70; 103; 105 (wón’th); 129 (And-Enge.lónd); 139. L. 127 has an additional weak syllable at the beginning (An’-a-thóusand).

7 (ecta). Ex. Crist-lát’-us év’re só to-dó; 17. So also 55; 136; 143.

7b (ecte). Ex. And-óv’ral mád’-hem fór to-cálle; 38. So also 53; 88 (rid’) ; 135 (strónglik.è).

8 (ecct). Ex. And-hwó-so díd-e wíduen wróng; 79.

8b (eccc). Ex. To-wrónge mícht-him nó-man brínge; 72. So also 83; 95; 96; 131.

9 (taaa). Ex. Hér’ y-shál bigínn’ a-rým; 21. So also 116 (com’); 123.

9b (taae). Ex. Óf a-tál’ ich-yóu wil-télle; 3. So also 102 (Fór to-hav’); 104.

9be (teae). Ex. And þe-tále ǁ ye-mów’ y-lére; 12. Cf. 113 (spék’ for speke); 125 (Shó-n’).

10 (tact). Ex. þann’ him-tók an-ível stróng; 114.

10b (taea). Ex. þát may-ríd’ on-áni stéde; 26. So also 71; 85; 93; 97; 120; 126.

11 (teta). Ex. þát we-móten cóm’ him-tó; 18. So also 31, 32; 48; 62; 73; 75 (read þe-fád’rles); 122; 142.

11b (tete). Ex. pát ye-mówen nóu y-hére; 11.

{xxxv}
12 (tect). Ex. Ben’.dicamus Dómi-nó; 20. So also 22 (taking wél-god together); 110 (read n’hdv’d’-he); 111; 154.

12b (tecc). Ex. Thát he-shólden cómen swípe; 140. So also 146 (nó-met’); 156.

13 (ctaa). Ex. Sóne sáys’ intil his-hónd; 251. (Not common.) So 462.

13b (ctae). Ex. Lóverd hwát shal-mé to-réde; 118. Cf. 141; 299.

14 (ctet). Ex. Wér’-it clérc or-wér’-it kníht; 77. Cf. 331.

14b (ctec). Ex. Háv’lok was a-fúl-god góme; 7. So also 42; 54 (Óv’ral); 101; 150; 197.

15 (ccta). Ex. Whó-so déd’-hem wróng or-láth; 76. Cf. 80.

15b (ccte). Ex. púrh-ut Énglond wíth her-wáre; 52. Cf. 33 (prest’s).

16 (ccct). Ex. Hérkneth tó-me góde men; 1. Cf. 171. Line 2 is of the type 16e; i.e. it has and-állè in place of a ‘cadence,’ after the caesural pause.

16b (cccc). Ex. þanne míhte chápmen fáre; 51. Cf. 63 (Énge.lóndes); 66 (n’óþ’re) ; 98 (só-god); 121.

§ 20. A few lines present exceptional difficulties. Thus l. 13 seems to run thus: Át-þe bigínning of-úr-e tále; where biginning occupies the place of a ‘cadence’ only, owing perhaps to the caesural pause. But the line would be less heavy if we were either to omit þe or to read ginning. A few other difficult lines are discussed in the Notes.

Considered as a whole, the metre is well varied, and sufficiently good. If the reader finds the discrimination of types too troublesome, he has only to neglect the distinctions between them, and to fall back upon the old formulae bAbAbAbA, or AbAbAbA, to one or other of which (neglect-{xxxvi}ing the feminine rhymes) almost every line in the poem can be ultimately reduced.